in : A Sociological Analysis∗

Duane Edwards† April 22, 2016

Abstract

Guyana is currently listed among the five countries internationally with the highest suicide rates. The traditional sociological theories are inadequate when used to assess the high rate of suicide in Guyana. Data on suicide in Guyana goes against the dominant sociological explanations in every way possible. Against Durkheim’s solidarity theory, suicide occurs predominantly among the group with the highest levels of social, political and religious soli- darity. And against the imitation and diffusion theories of suicide, the suicide rate is largely restricted among one ethnic group in Guyana. This paper takes a sociological approach to the study of suicide in Guyana by examining data collected by the Ministry of Health. The data is used to question the theoretical accuracy and applicability of dominant theories of suicide in relation to Guyana. It also analyses the historical and cultural specificities and peculiarities of the three dominant ethnic groups in Guyana and concludes that the high rate of suicide among one ethnic group results from a combination of social and cultural factors. The social factors have to do with the specific way in which the ethnic subsystem relates with the larger social system, while the cultural factors relates to intraethnic cultural prescriptions which are facing strain from the larger cultural system.

Keywords: Suicide in Guyana, Theory of Suicide, Durkheim, Tarde,

∗This paper was prepared for presentation at Guyana’s Golden Jubilee Symposium Series, May 2016, Georgetown, Guyana †Duane Edwards is currently a graduate student at the University of the West Indies, Cave Hill Campus, Barbados. Email: [email protected] 1 Introduction

For the year 2012, the World Health Organisation lists Guyana as the country with the highest suicide rates out of over 150 countries. This is definitely not good news for a small country with a small population and with hardly the financial and institutional resources to combat such a public epidemic. The other countries with significantly high suicide rates are located in Africa and Asia. In fact, Guyana and are the only two countries in the Caribbean region with suicide rates above the 15 per 100,000 threshold. In the words of the Director General of the World Health Organisation (World Health Organization, 2014), this level of suicide in any country is enough to brand suicide a public health problem and to occasion studies into its causes and consequences. This paper seeks to make its modest contribution towards understanding this problem in Guyana. In so doing it subjects data on suicide in Guyana to a sociological analysis utilising the dominant theoretical frameworks in suicidology.

This paper is divided into three main sections. The first section gives brief exposition of the dominant theories of suicide in sociology. The theories of Durkheim and Tarde will be explored with the hope that they can tell us something about suicide in Guyana. In the following section, data on suicide in Guyana will be used to test the validity of the hypotheses generated from the theories of suicide provided by Durkheim and Tarde. And, in the last section the social and cultural systems in Guyana will be subjected to a critical analysis with the hope that such an analysis can reveal more about suicide in Guyana than what the theories of Durkheim and Tarde can.

2 Durkheim’s Theory of Suicide

In his classic study of suicide, Durkheim proposed several hypotheses relating to the rela- tionship between social integration and suicide. In this study Durkheim developed a typology of suicide and argued that different types of have different relationships with social solidarity and social regulation. and egotistic suicide, he argues, varies pos- itively and negatively with social solidarity. Another type of suicide identified was anomic suicide which varies negatively with social order (Durkheim, 1951). The interesting thing about Durkheim’s effort is that it leaves us with testable hypotheses amenable to quanti- tative analysis. In fact, Durkheim had tested his hypotheses with data culled from French departments in his days the result of which led to the confirmation of his hypotheses.

1 It is interesting to note that Durkheim’s theory of suicide represents a special case of his general sociology. Arguing about the rules of sociological method, Durkheim insists that the proper concern of sociology should be with social facts (Durkheim, 1982). He distinguishes social fact from psychological and other facts and sets aside social facts as the subject matter of sociology. Social facts he argues are those facts which though external to us have tremendous influence on the way we act as social beings. Durkheim’s social facts are like Berger’s first order social institutions which impose themselves on us by their very facticity; they need no external sanctions (Berger and Luckmann, 1991). Economic exchange can be used as a good example of a social fact. Exchange via money as a legal tender and as a medium of exchange is an institutionalized reality. If one intends to make a purchase with leaves from a tree or some other material the exchange will not occur and the person will not get what s/he wants. If the thing s/he wants is highly needed, then s/he will have to find some way of getting money in order to effect the purchase. Let’s say s/he tries to get it by some other means such as printing counterfeit money or thieving it, s/he will then have to face the sanctions of the second order institutions if s/he gets caught. So, while first order institutions impose themselves by their very facticity, second order institutions require sanctions. Durkheim’s social facts are therefore first order institutions, and social integration and regulation are examples of social facts. The nature of social facts in any society influences other phenomena in those societies such as birth and death rate, the level of insanity, political and economic issues. His theory of suicide then is a specific example of how the nature of social facts in various countries influences the rates of suicide in those countries.

3 Tarde’s Theory of Suicide

The other dominant sociological theory of suicide evolves from Tarde’s argument that imitation rather than social fact is a more important social phenomenon and can better account for the clustering of suicide than social integration can. .

Similar to Durkheim, Tarde’s theory of suicide is a special case of his general sociology. Tarde argues that as in the natural sciences, the social sciences should consist in ’...view- ing any fact whatsoever under three aspects, corresponding, respectively, to the repetitions, oppositions, and adaptations which it contains, and which are obscured by a mass of vari- ations, dissymmetries, and disharmonies’ (De Tarde, 1899, 9). Repetition, the first link in these three important subject matter has been the focus of Tarde’s sociology. For him social phenomena develop as a result of social repetition. This means that the acts of social being

2 Regions 2003 2004 2005 2006 2007 Total 1 0 0 1 3 3 7 2 20 22 16 13 19 90 3 34 12 8 35 38 127 4 58 80 84 53 45 320 5 20 4 7 17 12 60 6 60 66 50 74 63 313 7 3 0 2 3 0 8 8 1 0 0 0 0 1 9 2 0 1 0 1 4 10 3 2 2 4 5 16 Total 201 186 171 202 186 946 Table 1: Frequency table of suicide by regions between the years 2003 - 2007 should be interpreted not on account of the facticity that social facts exercise over their actions but on account of the social influence that the actions of others have over their own actions (De Tarde, 1899).

If Tarde is correct, imitation effect should produce clustering above and beyond the cluster- ing of social integration (Baller and Richardson, 2002). This means that contiguous regions should show similar patterns of suicide whether or not those regions have similar levels of social integration. Different ethnic groups living in close proximity in one region should also display similar trends in suicide rate but different from the same ethnic group living far apart.

4 A Test of Both Theories

Suicide in Guyana, a country which ranks among the highest in global suicide rates, will be analysed using the theoretical frameworks of Durkheim and Tarde and the hypotheses generated from these frameworks. This will be done to see which of these theories provide a better understanding of suicide in Guyana. Towards this end, two specific hypotheses will be tested; the first generated from Durkheim’s theory and the other generated from Tarde’s theory. Durkheim’s Hypothesis Suicide rates vary inversely with social integration

4.1 Data on suicide in Guyana

The data provided in Table 1 and 2 was made available by the Stats Unit of the Ministry of Health, Guyana. It covers the period 2003 - 2007.

Social Integration in Guyana The figures of social integration have been compiled from various sources. Political integration has been compiled from the results of the 2006 national . An entropy

3 Ethnicity 2003 2004 2005 2006 2007 Total African 12 18 15 21 12 78 Amrindian 11 2 3 5 10 31 East Indian 156 153 140 161 155 765 Portuguese 1 0 1 0 0 2 Other 16 13 12 14 9 64 Unknown 5 0 0 1 0 6 Total 201 186 171 202 186 946 Table 2: Frequency table of suicide by ethnicity between the years 2003 - 2007

Regions 1 2 3 4 5 6 7 8 9 10 Suicide rate 33.86 188.80 126.47 106.01 124.10 252.38 89.55 89.79 66.75 65.07 Political Integration 1.57 1.25 1.21 1.57 1.16 1.09 2.02 2.26 2.24 1.57 Social Integration 27.2 35.38 22.94 26.47 21.00 33.20 16.44 26.95 22.72 30.99 Religious Integration 0.54 1.40 1.49 1.17 1.44 1.44 0.95 0.84 0.13 0.68 Table 3: Table of suicide rate and various measures of social integration by regions test was carried out on the data to determine the level of voting entropy in each region. The figures on social integration has been compiled from the Poverty Reduction Strategy paper by the Government of Guyana. And the figures on religious integration has been compiled from data provided by the National Census of 2002.

Both religious and political integration were calculated using Shannon’s entropy. For ex- ample in the case of political integration, if three (k) parties contest and one party dominates in a region that is also dominated by one ethnic group it means that entropy is low, hence conformity and political solidarity is relatively high in that region. The same was done for religious integration using data provided by the 2002 National. The formula for Shannon’s entropy is given as: k X H(X) = − pi log(pi) (1) 1 where k is the amount of political parties contesting election in the case of political integration and the amount of relgious organisations in the case of religious integration and p is the probability of these parties or religious organisations being the choice of the population. Entropy has been used by a number of researchers in the social sciences to measure conformity and segregation (Edwards, forthcoming 2016; Mora and Ruiz-Castillo, 2011; Coleman, 2004).

The figures in table (3) show that the regions with the highest rates of suicide are also the regions with the highest rates of social solidarity. This goes against the Durkheimian hypothesis that the level of egotistic suicide is negatively related to social integration. To be fair to Durkheim, however, it should be made known that Durkheim also hypothesized that altruistic suicide is positively related to social integration. This means whether integration is high or low one is expected to experience a certain level of suicide in a country. While the suicide rate might remain constant, it is the types of suicide which changes based on the

4 levels of social solidarity in a country. Having, therefore, established by means of empirical data that the highest levels of suicide occurs among the group with the highest level of social solidarity, could we assume that the kind of suicide we experience in Guyana is not what was formerly assumed to be egotistic suicide, but is indeed altruistic suicide?

Durkheim’s definition of altruistic suicide removes any ascription of intentionality from the perpetrator of altruistic suicide. As opposed to egotistic suicide where the decision to end one’s life rest entirely up to the individual, with altruistic suicide, the decision is directly imposed on the individual by society. It is for this reason that Durkheim insist that the label ’obligatory’ should precede altruistic suicide (Durkheim, 1951). Durkheim concedes that there are some varieties of altruistic suicide which are not as socially obligatory as the ’obligatory altruistic suicide’. What distinguishes non-obligatory kinds of altruistic suicide from egotistic suicide however, is that while in the case of egotistical suicide society frowns on such acts, it reacts favorably to the acts of altruistic suicides. The kinds of suicide encountered in Guyana falls no doubt under the egotistic category outlined by Durkheim (Edwards, forthcoming 2016; Danns, 2004).

Having been tested on suicide in Guyana, Durkheim’s hypothesis will now be tested using global data on suicide rates and ethnic and religious fractionalization and polarization as indices of social distance. The figures below provide a brief description of the data. The first descriptive figures apply to suicide and the second to social fractionalization (being used here as the inverse of social integration).

Min 1st Qu Median Mean 3rd Qu Max 0.400 4.700 7.800 8.901 11.500 44.200

Min 1st Qu Median Mean 3rd Qu Max 0.0200 0.5250 0.6756 0.8900 1.0000

In a study of the relationship between suicide and social integration among 113 countries, it was clear that there was no significant relationship between social integration and suicide in those countries. Suicide rates did no go up or down with the variation in the levels of social integration. The suicide data used in this analyse was extracted from the WHO database for suicide rates in 2012. The data used for social integration was extracted from dataset used in Reynal-Querol (2002). The two datasets were merged and entries with empty values were omitted. This resulted in a complete dataset of 113 countries for which there were values for both suicide and social integration.

5 Figure 1: Scatterplot of suicide and social distance

6 Figure 2: Scatterplot of suicide and social distance tau = 0.95

While the trendline in figure (1) is slightly positively sloped, it shows no significant relation- ship between the two variables in question. This disconfirms, at a global level, Durkheim’s theory of suicide. If one takes a closer look at the graph, however, it is evident that most of the countries with rates of suicide beyond 15 per every 100,000 are in the high social distance category. It means that most countries with very high levels of suicide rate are also countries with very low levels of social integration. The suicide rate of 15 was not arbitrarily selected; coincidendally, it was selected by means of the peak over threshold method in extreme value analysis, but it is also the level used by the WHO as representing a high rate of suicide. If we, therefore, draw a horizontal line at this threshold and perform a quantile regression analysis on the observations above this threshold, the relationship between the two variables show a relatively strong positive relationship (that is between suicide rates and social distance). It means that although Durkheim’s theory does not apply to countries with suicide rate below 15, it probably applies to countries with very high suicide rates. It is only at the 95 percentile that a significant positive relationship between suicide and social integration is shown.

7 Both at the national and global level, therefore, the data provided seems to support the point made by Edwards (forthcoming 2016) that although it is necessary as a condition social integration cannot be said to be the cause of suicide. Suicide therefore has to be studied in a way which takes into account factors other than the rise and fall in social solidarity as proposed by Durkheim. The fact that there are countries in the low social integration scale with low suicide rates also means that low social integration is simply a necessary condition, though not a immediate cause of high rates of suicide.

Tarde’s Hypothesis Baller and Richardson (2002) argue that if imitation effect is present in suicide rates they should manifest themselves in geographic patterning of suicide. Suicide rates, therefore, should cluster geographically even when social integration varies. This means that in Guyana’s context, regions with high rates of suicide should be contiguous and that groups living close to each other should show similar suicide rates based on the imitation effect. Figure 3 rep- resents a geographic mapping of suicide in Guyana by regions. The data used to generate this map is displayed in Table 3. Darker regions are regions which are very high in suicide rates, while lighter regions are those with low suicide rates. The figure shows that the two regions with the highest rates of suicide, namely regions 2 and 6, are regions which are remote from each other. Moreover, regions contiguous to those regions show varying levels of suicide rates. For region 2, the regions near by are regions 1 and 3. While region 3 shows relatively high suicide rates region 1 shows significantly low suicide rates. The same goes for region 6. While region 5 shows relatively high suicide rates, region 7 shows significantly low rates. More importantly, however, is the fact that suicide rate are concentrated among one ethnic group even in cases where those groups live in regions far removed from each other.

A spatial analysis of suicide in Guyana goes against the imitation hypothesis in two ways. First of all, though suicide seems to reflect geographic clustering, this clustering occurs among one ethnic group only. Secondly, the two regions with the highest suicide rates are two regions which are geographically remote from each other. The intervening space between these two regions are regions with comparatively lesser rates of suicide than the two extreme regions.

5 Suicide in Guyana: an Inductive Approach

Danns (2004) in his study of suicide in Guyana found that the two highest reasons for suicide in Guyana are relationship problems and family problems. Therefore suicide will be analysed taking these two main factors into consideration.

8 Suicide Map

Figure 3: Geographic mapping of suicide rates

5.1 Suicide caused by relationship problems

In societies characterised by high levels of social integration, or what Durkheim refers to as mechanical solidarity everyone is tied to what is referred to by Mead as the generalised other. This generalised other takes the foreground in such societies. When there is a transition to organic societies, it is not that this generalised other is totally eliminated, it is just that it takes a background role. As a result, persons make up for this lost by forging more strong and interdependent relationships with each other. Whenever there is a strain or disruption in these relationships, persons feel abandoned to the social void created by the lack of mechanical solidarity. It is important to note that this feeling of abandonment need not accurately reflect structural reality, it is enough that it exist only in the individual’s mind for it to have a destructive influence on the individual. This is where Parson's action theory, influenced by Weber's, is very insightful. By way of an example, let's go with the Parsonsian assertion that a social system is made up of a multiplicity of actors fulfilling a multiplicity of roles (Parsons, 1991). Therefore, at any one time, ego and alter, while interacting with each other in some cathetic relationship is also interacting with a multiplicity of others, being a part of various social networks. For example, an individual might have social relationships at any one time with family members (mother, father, siblings, cousins,

9 aunts, etc.), school friends, church friends, club associates, and lovers. But a disruption in one of the relationships, lets say the lovers relationship, would be enough to inspire in the individual a feeling of lost and abandonment and thereby drive the individual to commit suicide. This is so even if all the other relationships remain intact. This depends, however, on the emotional investment ego makes in that one relationship and the kinds of returns he is expecting from it. If the emotional investment is high, his expectations of alter will also be high, and if his emotional investment is low his expectations will be low. What this shows is that it is not necessarily the change in the macro situation that causes one to commit suicide, but merely the interpretation (or the social meaning) one gives to the disruption which occurs in the micro situation (Berk, 2006). While this example highlights the origin of Durkheims egotistic suicide, the same can be said for anomic suicide. Whenever, there is a financial crisis, it is not the crisis itself which causes the millionaire to commit suicide, but the meaning he attaches to living at a lower level of wealth or living in poverty.

From this standpoint all of Durkheims types of suicide look interestingly similar. Although they are all occasioned by different external circumstances, they result from very similar in- ternal motivation. Egotistic suicide is caused by ego attaching a certain meaning to the break that occurs in the relationship with himself and alter; anomic suicide by the meaning (fear, apprehension, etc) he attaches to living in an anomic situation; altruistic suicide by the meaning he attaches to his relationship with alter; and fatalistic suicide by the fear that he has no control over the prevailing situation. In all these cases, Durkheim was correct when he highlighted the social conditions necessary for these types of suicides to occur. He erred, however, when he attributed to them final causality. It does not take much thinking to conclude that were these factors sufficient causes their very existence across the entire social network would have given rise to a far higher incidence of suicide within the network. Having said the above, the question still remains what are the specific elements in the cultural sub- system of Indo-Guyanese that is so different from other ethnic groups to the extent that the presence of those elements influences suicidal behaviour among them? Speaking specifically for suicide which has been influenced by relationship problems, the cultural elements have to do with the level of sexual freedom among the Indo-Guyanese. As already demonstrated by the data provided, a higher level of mechanical solidarity exists among the Indo-Guyanese group than among the other ethnic groups. Let’s take the Afro-Guyanese group as an exam- ple. This group possesses relatively high sexual freedom partly because of the high levels of organic (as opposed to mechanical) solidarity which prevails within the group (Matthews and Wilson, 1999; Landis, 1973). As a result, it is expected that members of this groups would be more opened to having loose sexual relationships and therefore reacts differently when

10 cheated upon by their partners. For example, there is a saying among the AfroGuyanese group that a man should expect two things in life: ’blow and goadie’. Blow refers to sexual and marital infidelity, while goadie is the folk term for hydrocele. The meaning and norma- tive structure from which this saying emerges psychologically prepares the Afro Guyanese actors to deal with any act of infidelity by their partners. Another cultural pattern within the AfroGuyanese group which plays a significant role in minimising the suicidogenic ten- dencies among them is the fact that they marry at a relatively later age and they are less conservative than their Indo Guyanese counterpart (Samuel and Wilson, 2008). This gives them the opportunity of being part of several loose relationships before they finally settle down in a more committed one at an older age.

These two cultural factors pinpointed above are relatively absent from the Indo Guyanese group whose members both marry at a relatively younger age and lack the cultural values which trivialise sexual and marital infidelity. As a result they are less able to cope with disequilibrium within their sexual relationships. The cultural factors outlined above combined with alienative needdisposition in the personality subsystem all work together to influence high rates of egotistic suicides among IndoGuyanese actors. However, all these work against a backdrop of low social solidarity in the larger social system.

5.2 Suicide caused by Family Problems

One of the main differences between the IndoGuyanese ethnic group and other ethnic groups in Guyana is that the IndoGuyanese group subscribes to values and norms that are inconsistent in some ways with the norms and values of the larger social system (Samuel and Wilson, 2008; Smith, 2001). This creates ambivalence among IndoGuyanese actors who are always called upon to either acquiesce to the demands of the ethnic subsystem or to those of the larger social system.

The IndoGuyanese ethnic group has always been a more closed group than other groups in Guyana. The reasons for this are twofold. In the first instance, the lack of openness results from their relationship with the plantation system which has been theorised as a closed system (Smith, 2001; Knottnerus et al., 1999). Unlike the Africans whose relationship with this system was outwardly hostile, antagonistic, and repressive and who moved away from the system in order to be incorporated in the colonial society, the IndoGuyanese who replaced the Africans had a more settled relationship with the plantation system and developed their communities within and around the already established total structures.

11 The other reason is that IndoGuyanese were introduced into the Guyanese creole society at a relatively later date than most other ethnic groups; this resulted in a comparatively later exposure to modern universalistic values which were incorporated into the creole system. These two factors helped to reinforce family norms based on values which were at odds with the values of the larger society. One expression of this is the norm of collectivity (over that of self) expressed in the need by the family to dictate to the individual issues relating to sexual relationships. In an effort to maintain or augment the family status, families dictate that their members must marry persons within their own race (or higher) and class (or higher). This normative element takes away from the individual the freedom to make decisions relating to the most personal aspects of their lives. Landis (1973) captured this sentiment in an empirical research he conducted in his analysis of racial attitudes among Africans and Indians in Guyana. According to him, 'four-fifth of the Indian respondents said they would object to their son or daughter marrying an African. Only 16 per cent of the Africans said they would object to their son or daughter marrying and Indian'. This quest for control by parents in an era of individual freedom creates tension in the parent-child relationship. As a reaction to this control, children often choose to rebel by committing suicide.

Although racial issues are not the only issues for which parent and children come into sharp conflict, they are enough to demonstrate the disruption in the parent child relationship and the pressure created when traditional values defended by the parent comes into conflict with modern values adopted by the child. The first point to emphasize is that the ways in which pressure is exerted on the motivational system of the actor will vary as a function of the kind of pattern with which he is expected to conform. (Parsons, 1991, 181). This paper takes the position that the normative pattern in the IndoGuyanese community places tremendous pressure on IndoGuyanese actors, especially the younger generation, to conform. The nor- mative pattern of the larger social system which runs contrary to the ethnic subsystem, also places pressure on the actor, thereby forcing actors to commit suicide as a form of rebellion against the imposition of the family and community.

6 Conclusion

Suicide in Guyana, it is clear, does not fit into the two established theoretical frameworks in sociology. It therefore must be studied inductively so as to get a sense of the peculiar- ities of this phenomenon in Guyana. The data on suicide in Guyana when subjected to a Durkheimian analysis goes against the Durkheimian hypothesis in many ways. It is the same

12 for Tarde’s theory of imitation. Suicide in Guyana does not exhibit any imitation effect as the highest rate are concentrated among one ethnic group. This ethnic group shows a high tendency to commit suicide irrespective of which geographical location they live in. More- over, suicide seems not to have a contagion effect on groups which live in close proximity to this ethnic group. When studied inductively, however, suicide in Guyana reveals many interesting facts which have consequences for the development of a more robust theory of suicide as proposed by Edwards (forthcoming 2016).

References

Baller, R. D. and Richardson, K. K. Social integration, imitation, and the geographic pat- terning of suicide. American Sociological Review, pages 873–888, 2002.

Berger, P. L. and Luckmann, T. The social construction of reality: A treatise in the sociology of knowledge. Number 10. Penguin UK, 1991.

Berk, B. B. Macro-micro relationships in durkheim’s analysis of egoistic suicide. Sociological Theory, 24(1):58–80, 2006.

Coleman, S. The effect of social conformity on collective voting behavior. Political analysis, 12(1):76–96, 2004.

Danns, G. K. The shadow of death, 2004.

De Tarde, G. Social laws: An outline of sociology. The Macmillan company, 1899.

Durkheim, E. Suicide: a study in sociology [1897]. Translated by JA Spaulding and G. Simpson (Glencoe, Illinois: The Free Press, 1951), 1951.

Durkheim, E. The rules ofthe sociological method, 1982.

Edwards, D. Suicide in guyana: A parsonsian corrective to durkheim’s theory of suicide. Canadian Journal of Latin American and Caribbean Studies, 41,3, forthcoming 2016.

Knottnerus, J. D., Monk, D. L., and Jones, E. The slave plantation system from a total institution perspective. Plantation society and race relations: The origins of inequality, pages 17–27, 1999.

Landis, J. B. Racial attitudes of africans and indians in guyana. Social and Economic Studies, pages 427–439, 1973.

13 Matthews, L. and Wilson, L. C. Ethnic tolerance in urban guyana: A conceptual and empirical analysis. International review of modern sociology, pages 55–75, 1999.

Mora, R. and Ruiz-Castillo, J. Entropy-based segregation indices. Sociological Methodology, 41(1):159–194, 2011.

Parsons, T. The social system, 1991.

Reynal-Querol, M. Ethnicity, political systems, and civil wars. Journal of Conflict Resolution, 46(1):29–54, 2002.

Samuel, P. S. and Wilson, L. C. Values and beliefs of indo-guyanese: An assessment of the assimilation hypothesis. Man in India, 88(1):83, 2008.

Smith, R. T. Social stratification, cultural pluralism and integration in West Indian societies. 2001.

World Health Organization. Preventing suicide: A global imperative. World Health Organi- zation, 2014.

14